/Patent Tracker
SamsungPatent applicationAI & softwareInterest 6/10

Samsung patents method to track sound objects in video using AI

Patent drawing
Patent drawing from the official filing.

Samsung has filed a patent for a system that uses a neural network to track specific sounds, like a voice or a car horn, across different moments in a video. The system takes a known sound from one frame and predicts where that same sound appears in the next frame, allowing for continuous tracking of individual audio elements.

Why it mattersThis technology could improve how devices separate and manipulate individual sounds in a scene, which is useful for features like isolating a speaker's voice from background noise or enhancing audio in video editing.

In the filing’s own words

An electronic apparatus includes at least one processor and memory storing instructions, where the instructions, when executed by the at least one processor individually or collectively, cause the electronic apparatus to obtain at least one first audio object mask corresponding to a first audio frame among a plurality of audio frames, identify at least one second audio object mask corresponding to a second audio frame that is after the first audio frame by inputting the obtained at least one first audio object mask into a first neural network model and identify at least one second audio object in the second audio frame based on the identified at least one second audio object mask.

What exactly they claim (claim 1)

1. An electronic apparatus comprising: at least one processor; and memory storing instructions, wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic apparatus to: obtain at least one first audio object mask corresponding to a first audio frame among a plurality of audio frames; identify at least one second audio object mask corresponding to a second audio frame that is after the first audio frame by inputting the obtained at least one first audio object mask into a first neural network model; and identify at least one second audio object in the second audio frame based on the identified at least one second audio object mask.

Facts

Company
Samsung (SAMSUNG ELECTRONICS CO., LTD.)
Official title
ELECTRONIC APPARATUS FOR SEPARATING AUDIO OBJECTS INCLUDED IN AN AUDIO SIGNAL, METHOD FOR OPERATING THEREOF AND STORAGE MEDIUM
Publication number
US20260331128A1
Application number
19631029
Filed
Mar 27, 2026
Published
Oct 22, 2026
Inventors
3
Claims
20

Classified under

G10L 25/84 Physics & computingG10L 17/02 Physics & computingG10L 21/0272 Physics & computingG10L 25/30 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Samsung in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.